A case study on frontal-face images
Francisco de Assis, Pereira Vasconcelos de Arruda, José Eustáquio, Rangel de Queiroz, Herman Martins Gomes · 2012
We present and evaluate a neural network-based technique to automatically enable NPR renderings from dig- ital face images, which resemble semi-detailed sketches. The technique has been experimentally evaluated and com- pared with traditional approaches to edge detection (Canny and Difference of Gaussians, or DoG) and with a more re- cent variant, specifically designed for stylization purposes (Flow Difference of Gaussians, or FDoG). An objective evaluation showed, after an ANOVA analysis and a Tukey t- test, that the proposed approach was equivalent to the FDoG technique and superior to the DoG. A subjective experiment involving the opinion of human observers proved to be com- plementary to the objective analysis.